fix(cron-tool): use generic object schema for job/patch to fix Claude via Antigravity by mukhtharcm · Pull Request #280 · openclaw/openclaw

mukhtharcm · 2026-01-06T07:06:52Z

Summary

Fixes a critical bug where the cron tool fails when using Claude models via Google Cloud Code Assist (Antigravity) with:

tools.9.custom.input_schema: JSON schema is invalid. It must match JSON Schema draft 2020-12

This is a partial revert of #256 - keeps the valuable normalization logic (normalizeCronJobCreate/normalizeCronJobPatch) but reverts the schema change that broke Antigravity+Claude.

Problem

PR #256 changed the cron tool schema from:

job: Type.Object({}, { additionalProperties: true })

to:

job: CronAddParamsSchema

This broke Claude via Antigravity because of a complex interaction between TypeBox, pi-ai schema sanitization, and Claude's JSON Schema validation.

Root Cause Analysis

1. CronAddParamsSchema contains nested Type.Union

// In gateway/protocol/schema.ts
export const CronScheduleSchema = Type.Union([
  Type.Object({ kind: Type.Literal("at"), atMs: Type.Integer() }),
  Type.Object({ kind: Type.Literal("every"), everyMs: Type.Integer() }),
  Type.Object({ kind: Type.Literal("cron"), expr: NonEmptyString }),
]);

export const CronPayloadSchema = Type.Union([
  Type.Object({ kind: Type.Literal("systemEvent"), text: NonEmptyString }),
  Type.Object({ kind: Type.Literal("agentTurn"), message: NonEmptyString, ... }),
]);

2. TypeBox compiles Type.Union to JSON Schema `anyOf`

{
  "job": {
    "properties": {
      "schedule": {
        "anyOf": [
          { "type": "object", "properties": { "kind": { "const": "at" }, "atMs": { "type": "integer" } } },
          { "type": "object", "properties": { "kind": { "const": "every" }, "everyMs": { "type": "integer" } } },
          { "type": "object", "properties": { "kind": { "const": "cron" }, "expr": { "type": "string" } } }
        ]
      }
    }
  }
}

3. pi-ai's `sanitizeSchemaForGoogle()` strips `anyOf`

The schema sanitizer removes unsupported JSON Schema keywords for Google/Gemini APIs:

const unsupportedKeywords = [
  'patternProperties', 'const', 'anyOf', 'oneOf', 'allOf', 'not',
  // ... more
];

After sanitization, nested anyOf properties become empty objects:

{
  "job": {
    "properties": {
      "schedule": {},
      "payload": {},
      "sessionTarget": {},
      "wakeMode": {}
    }
  }
}

4. Claude rejects empty `{}` schemas

Claude requires valid JSON Schema 2020-12 with explicit type fields. An empty schema {} has no type, which Claude rejects:

tools.9.custom.input_schema: JSON schema is invalid.
It must match JSON Schema draft 2020-12

Why This Wasn't Caught

Provider-specific bug: Only affects Claude via Antigravity, not direct Claude API or Gemini
Recent change: PR Cron: normalize cron.add inputs + align channels #256 was merged Jan 5, 2026
Complex interaction: Requires understanding TypeBox → JSON Schema → sanitization → Claude validation chain
Gateway still worked: The gateway has its own validateCronAddParams() that works fine

Solution

Revert job/patch schemas to Type.Object({}, { additionalProperties: true }):

// Before (broken)
job: CronAddParamsSchema,
patch: CronJobPatchSchema,

// After (fixed)
job: Type.Object({}, { additionalProperties: true }),
patch: Type.Object({}, { additionalProperties: true }),

This produces valid JSON Schema:

{ "job": { "type": "object", "additionalProperties": true } }

Why This Is Safe

Runtime validation preserved: normalizeCronJobCreate() and normalizeCronJobPatch() still clean up AI inputs (this is the valuable part of Cron: normalize cron.add inputs + align channels #256)
Gateway validation preserved: validateCronAddParams() provides clear error messages for invalid job structures
AI still works: Claude can infer the job structure from the tool description and context. The lack of explicit schema doesn't prevent it from creating valid jobs.
Matches working version: This is exactly what was running before Cron: normalize cron.add inputs + align channels #256 and worked fine.

Testing

✅ Tested with claude-opus-4-5-thinking via Google Antigravity
✅ Schema error no longer occurs
✅ Cron job creation works correctly
✅ Build passes

Future Considerations

The proper fix would be to enhance pi-ai's sanitizeSchemaForGoogle() to handle anyOf more gracefully:

Merge anyOf variants into a single object with optional properties
Or convert anyOf of literals to enum
Or add a type field when stripping anyOf leaves an empty schema

This would allow using rich schemas while maintaining Claude compatibility. Filed as a consideration for upstream.

… via Antigravity ## Problem When using Claude models via Google Cloud Code Assist (Antigravity), the cron tool fails with: tools.9.custom.input_schema: JSON schema is invalid. It must match JSON Schema draft 2020-12 ## Root Cause Analysis 1. **CronAddParamsSchema contains nested Type.Union** - schedule: Type.Union([at, every, cron]) - payload: Type.Union([systemEvent, agentTurn]) - sessionTarget: Type.Union([main, isolated]) - wakeMode: Type.Union([now, next-heartbeat]) 2. **TypeBox compiles Type.Union to JSON Schema anyOf** ```json { "schedule": { "anyOf": [ { "type": "object", "properties": { "kind": { "const": "at" }, ... } }, { "type": "object", "properties": { "kind": { "const": "every" }, ... } }, ... ] } } ``` 3. **pi-ai sanitizeSchemaForGoogle() strips anyOf** The sanitizer removes unsupported JSON Schema keywords for Google APIs, including anyOf, oneOf, allOf, const. After sanitization: ```json { "schedule": {} } ``` 4. **Claude rejects empty {} schemas** Claude requires valid JSON Schema 2020-12 with explicit types. An empty schema {} has no type field, which Claude rejects as invalid. ## Why This Worked Before Commit a322075 (Jan 5) used Type.Object({}, { additionalProperties: true }) for job/patch, which produces { "type": "object" } - valid for Claude. Commit 67e1452 (PR openclaw#256, Jan 5) changed this to CronAddParamsSchema to provide better schema hints to the AI. This broke Antigravity+Claude. ## Solution Revert job/patch schemas to Type.Object({}, { additionalProperties: true }). This is safe because: - Runtime validation via normalizeCronJobCreate/normalizeCronJobPatch (kept) - Gateway validates via validateCronAddParams with clear error messages - The AI can still create valid job objects from the tool description ## Testing - Tested with Claude opus-4-5-thinking via Antigravity - Schema error no longer occurs - Cron job creation still works correctly Partial revert of 67e1452 - keeps normalization logic, reverts schema change.

* Add execution layer runtime parity gap analysis Comprehensive analysis of Pi Runtime vs Claude Agent SDK feature gaps in the unified execution layer, with 20 prioritized next steps. https://claude.ai/code/session_017oEzmayzdirGAKmSw2ryQZ * Meridia: wire multi-factor scoring into capture hook * merge/minor fixes for ui/* * Meridia: add per-capture graph fanout queue with retries * Meridia: enforce sanitization before persistence and fanout * Meridia: complete Tier2 vector probing and Postgres vector support --------- Co-authored-by: Claude <[email protected]>

* feat: tool journal/diagnostics * feat: journal fixes * feat(ui): add error boundary component with retry & friendly messages - New error-boundary.ts component with renderError/renderErrorIf helpers - Custom element <error-boundary> with auto-retry and exponential backoff - friendlyError() maps raw errors to user-friendly messages + suggestions - Supports severity levels (danger/warning/info), compact mode, dismiss - Collapsible technical details section - ARIA compliance with role=alert and aria-live - Replaces all inline callout danger patterns across 23 view files - Consistent error UX across agents, channels, sessions, config, etc. * Web: reset retry timers on error changes (openclaw#273) * Gateway: unify exec approvals with tool approval flow (openclaw#319) * Gateway: unify exec approvals * Gateway: guard exec approval resolves * Feat/pr review monitor (openclaw#313) * minor fixes * feat: monitor AI PR review comments * PR review monitor: add pagination config (openclaw#324) * Codex/review branch changes and identify issues (openclaw#325) * minor fixes * feat: monitor AI PR review comments * PR review monitor: add pagination config * UI: reset auto-retry timers on error changes (openclaw#328) * feat(ui): add error boundary component with retry & friendly messages - New error-boundary.ts component with renderError/renderErrorIf helpers - Custom element <error-boundary> with auto-retry and exponential backoff - friendlyError() maps raw errors to user-friendly messages + suggestions - Supports severity levels (danger/warning/info), compact mode, dismiss - Collapsible technical details section - ARIA compliance with role=alert and aria-live - Replaces all inline callout danger patterns across 23 view files - Consistent error UX across agents, channels, sessions, config, etc. * Web: reset retry timers on error changes (openclaw#273) * UI: reset auto-retry timers on error changes * Add execution layer runtime parity gap analysis (openclaw#280) * Add execution layer runtime parity gap analysis Comprehensive analysis of Pi Runtime vs Claude Agent SDK feature gaps in the unified execution layer, with 20 prioritized next steps. https://claude.ai/code/session_017oEzmayzdirGAKmSw2ryQZ * Meridia: wire multi-factor scoring into capture hook * merge/minor fixes for ui/* * Meridia: add per-capture graph fanout queue with retries * Meridia: enforce sanitization before persistence and fanout * Meridia: complete Tier2 vector probing and Postgres vector support --------- Co-authored-by: Claude <[email protected]> * Codex/review branch changes and identify issues (openclaw#325) * minor fixes * feat: monitor AI PR review comments * PR review monitor: add pagination config * Work queue: add heartbeat leases (openclaw#329) * fix: duplicate lines on main * Tools: clarify work_item refs and workstream (openclaw#332) Co-authored-by: Claude Opus 4.6 <[email protected]> * Config: clarify agents.list placement, accept agents.list in web import, and document guidance (openclaw#331) * Config: clarify agents.list validation * Web: tighten agents list import validation * Sessions: align label limits (openclaw#333) * Work queue: add work item refs support (openclaw#312) * Tests: update migration count * Tools: accept refs in work_item tool * Work queue: link Codex tasks to PRs (post GitHub comments) (openclaw#337) * Work queue: link codex tasks to PRs * Work queue: skip branchPrefix-only PR lookup * Claude/runtime orchestrator tools eu d uu (openclaw#327) * feat(agents): add runtime tool-approval orchestrator with approvals.tools config - Add approvals.tools config types + zod schema (enabled, mode, timeoutMs, policy, routing, classifier) - Create tool-approval orchestrator module (decision engine, param redaction, gateway integration) - Integrate orchestrator into before-tool-call wrapper path (runs after plugin hooks, before execution) - Add ToolApprovalBlockedError with stable machine-readable error shape - Add 90 tests covering all mode/decision/risk branches - Backward-compatible: no behavior change when approvals.tools is missing or disabled * feat: upgrade /approve and Discord handler to canonical tool approvals - /approve now queries tool.approvals.get for canonical records and resolves via tool.approval.resolve (with requestHash); falls back to legacy exec.approval.resolve when no canonical record is found - Discord handler listens for tool.approval.requested/resolved events and renders generic tool approval embeds for non-exec tools - resolveApproval prefers tool.approval.resolve when requestHash is cached, keeping legacy exec path for backward compatibility - Updated command description to 'tool approval requests' - Added shouldHandleToolApproval for canonical event filtering - Extended tests with canonical, legacy-fallback, and gateway-error scenarios * refactor: rename .clawdbrain → .openclaw and fix repo/domain references - Settings dir: ~/.clawdbrain → ~/.openclaw - Repo references: openclaw/clawdbrain → dgarson/clawdbrain - Domain: clawdbrain.bot → openclaw.ai - CLI command: clawdbrain login → openclaw login - 48 files changed across src/, docs/, apps/web/, ui/ * cron timeout fixes * feat(agents): wire tool approval context from config into tool creation path - Inject approvals.tools config into wrapToolWithBeforeToolCallHook context - Populate channel field from messageProvider via resolveGatewayMessageChannel - Wire callGatewayTool as the gateway call adapter for approval requests - Approval context is only constructed when approvals.tools exists and is enabled * fix: address review gaps in tool approval handler - Exec dedup: store canonical request for exec tools and defer embed creation by 200ms so the legacy mirror gets first shot; if the mirror never arrives, fall back to a generic tool embed (future-proofs against legacy event removal) - Extract sendToolApprovalEmbed to eliminate code duplication - Add buildApprovalCustomId / parseApprovalData generic aliases (same wire format, clearer naming for non-exec tool code paths) - Add alias identity tests * fix: minor tool approval request fixes * auto-reply/approval integration fix * include exec approval doc * fix: agent-runner-execution integration into auto-reply, executor/kernel fixes * more work on agent runner and memory/heartbeta integration * lots of tests resulting from unification of exec kernel; refactored * Redact arrays in approval helper * lancedb fixes * more fixes/test updates * fix: minor problem * fix: restore proper non-throwing session label truncation --------- Co-authored-by: Claude <[email protected]> * Tool approval/protocol cleanup (openclaw#334) * infra: consolidate tool approval types and clean protocol schema * infra: bridge tool approval routing config into forwarder * agents: enrich tool approval decision engine with config resolution and reason codes * test: update tool approval tests for protocol and decision engine changes * infra: consolidate tool approval types and clean protocol schema * infra: bridge tool approval routing config into forwarder * agents: enrich tool approval decision engine with config resolution and reason codes * test: update tool approval tests for protocol and decision engine changes * chore: conflict resolution * chore: checkou tfrom main * Codex/map paramssummary to exec command field (openclaw#342) * infra: consolidate tool approval types and clean protocol schema * infra: bridge tool approval routing config into forwarder * agents: enrich tool approval decision engine with config resolution and reason codes * test: update tool approval tests for protocol and decision engine changes * infra: consolidate tool approval types and clean protocol schema * infra: bridge tool approval routing config into forwarder * agents: enrich tool approval decision engine with config resolution and reason codes * test: update tool approval tests for protocol and decision engine changes * chore: conflict resolution * chore: checkou tfrom main * Tool approvals: preserve exec command * Codex/add web inbox for tool approvals (openclaw#339) * Web: add tool approval inbox support * Web: fallback approval resolution * Web: fall back to agent approvals when IDs differ (openclaw#263) * memclawd: scaffold phase 0 service foundation (openclaw#330) * memclawd: apply oxfmt * Memclawd: add client samples and align pipeline config * Codex/implement work item refs system d2mkjz (openclaw#344) * Tools: clarify work_item refs and workstream * Tests: update migration count --------- Co-authored-by: Claude Opus 4.6 <[email protected]> * Codex/review branch changes and identify issues kuj3uy (openclaw#343) * Tests: update migration count * Tools: accept refs in work_item tool * Work queue: add refs reindex command * Work queue: align refs migration and add refs-reindex CLI (openclaw#345) * Tests: update migration count * Work queue: move refs backfill to 004 migration --------- Co-authored-by: Claude Opus 4.6 <[email protected]> --------- Co-authored-by: Claude <[email protected]>

… via Antigravity (openclaw#280)

mukhtharcm force-pushed the fix/cron-tool-schema-antigravity branch 2 times, most recently from a531e88 to 9012c22 Compare January 6, 2026 07:51

thewilloftheshadow merged commit 42d1c24 into openclaw:main Jan 6, 2026
26 checks passed

zooqueen pushed a commit to hanzoai/bot that referenced this pull request Mar 6, 2026

fix(cron-tool): use generic object schema for job/patch to fix Claude…

2f95a4a

… via Antigravity (openclaw#280)

filipvalica mentioned this pull request Mar 29, 2026

[Bug] /v1/responses returns 403 missing scope operator.write with valid JWT #57331

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(cron-tool): use generic object schema for job/patch to fix Claude via Antigravity#280

fix(cron-tool): use generic object schema for job/patch to fix Claude via Antigravity#280
thewilloftheshadow merged 1 commit intoopenclaw:mainfrom
mukhtharcm:fix/cron-tool-schema-antigravity

mukhtharcm commented Jan 6, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

mukhtharcm commented Jan 6, 2026

Summary

Problem

Root Cause Analysis

1. CronAddParamsSchema contains nested Type.Union

2. TypeBox compiles Type.Union to JSON Schema anyOf

3. pi-ai's sanitizeSchemaForGoogle() strips anyOf

4. Claude rejects empty {} schemas

Why This Wasn't Caught

Solution

Why This Is Safe

Testing

Future Considerations

Related

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

2. TypeBox compiles Type.Union to JSON Schema `anyOf`

3. pi-ai's `sanitizeSchemaForGoogle()` strips `anyOf`

4. Claude rejects empty `{}` schemas